For several years now, I’ve helped to organize and moderate a discussion forum focused on knowledge graphs, knowledge modeling and related topics. The group began as an effort to help each other build personal knowledge graphs.
Two of our members from that effort – George Anadiotis and Ivo Velichkov – compiled and edited a multi-author book published in June 2023 titled Personal Knowledge Graphs: Connected thinking to boost productivity, creativity and discovery available on Amazon that explored the topic more systematically. A third member, Margaret Warren, contributed a chapter on personal image-centric knowledge graphs.
In 2023, we reorganized our group and are working toward the creation of a think tank that one of our co-founders Pete Rivett has named the Dataworthy Collective (See dataworthy dot org for more information). The name highlights the necessity of intentionally shaping, sharing and harnessing the power of trusted, reusable knowledge in an organic, human-led, machine-assisted process.
The Dataworthy Collective is a shared resource staffed with data, knowledge and content experts, many of whom are working on their own products. Our initial goals are to raise public awareness of the group’s purpose and capabilities, promote the products members are developing and encouraging others who are like minded to join.
The subject matter the group coalesces around is foundational, constantly changing and at the same time diverse. That combination of evolving depth and breadth is what attracts people who are curious about logical graph modeling interests, explorations and practical activities. Peers present to each other to obtain feedback and learn from each other. We also invite guests doing innovative things to present to the group.
Graphs from informal to formal
Knowledge graphs can be useful at various points in their evolution and level of completeness. Our guide to the decentralized web Gyuri Lagos is mostly interested in simple content-addressed associations of person to person and person to topic to content, for example. Gyuri anticipates positive and growing numbers of deepening interactions triggered by a critical mass of the right kinds of associations.
Another member, Open Group veteran Chris Harding, has designed a virtual data lake platform called Lacibus and has been exploring the possibilities of text embeddings and generative AI in knowledge graph-oriented topic extraction.
Others build graphs that are designed to be more expressive, specific and structured to begin with, for purposes such as asset management. Margaret Warren, for instance, runs ImageSnippets, a site that uses a simple ontology to organize and make photo and other image collections discoverable and reusable with explicit, standard semantic metadata.
Philippe Höij for his part is positioning his new visual semantic data product builder DFRNT for the impending convergence of knowledge management, data and software engineering. DFRNT’s focus is user-friendly data modeling that takes advantage of forms and familiar structured content standards such as JSON-LD in conjunction with TerminusDB and its headless content management system features.
Still other members bring a variety of skills and experiences to the mix:
- Flores Bakker is an enterprise architect at the Ministerie van Financiën in the Netherlands working on semantic standards-based tooling for web pages and ontologies
- Gevik Nalbandian is CEO of AI application integration platform InsightNexus
- Emeka Okoye is a knowledge engineer at CYMANTIKs and founder of OpenDataNG who’s behind the Land Portal Foundation’s knowledge graph
- Andrew Padilla runs his own information management consultancy and has a background as a senior software engineer at IBM
- Pete Rivett is a knowledge graph architect, ontologist and director of the Enteprise Knowledge Graph Foundation
- Larry Swanson is a content architect, marketer and podcaster who’s been tracking the latest advances in CMSes
- Jessica Talisman is an architect, ontologist and taxonomist with a background in online marketing and advertising
- Ilana Zane is a knowledge engineer at insurtech startup Harbor.ai
There are more from four different continents who join from time to time, depending on the meeting topic and their availability.
Outlook for 2024
Our Collective is currently in the process of building out an initial website and planning its next steps. We meet every Friday at 0700 Pacific, 1000 Eastern, 1500 UK and 1600 Central European time to discuss knowledge graph, hybrid AI and related trends, technologies, products and services. It’s an informal group focused primarily on open source technologies. If you’re interested in taking a look and hearing about the many different innovative and important things we’re involved with, please feel free to reach out to me via LinkedIn. I can add you to the mailing list.